The PRIIPs regulation comes into force on 1 January 2018 and requires each Packaged Retail and Insurance-based Investment Product (PRIIP) manufacturer to create and maintain a Key Information Document (KID) for each of the PRIIPs that they sell, and for this document to be given to retail customers at the point of sale. The aim is that these short and consumer-friendly KIDs will encourage efficient markets by helping investors to better understand and compare the key features, risk, rewards and costs of different PRIIPs.
The regulatory technical standards (RTS) provides practical details such as the KID must;
- follow the defined structure
- be at most three pages long
- be void of marketing information
- be in a readable font
- look okay when printed or photocopied copied in black and white
The RTS also provides detailed technical information as to the content presented within the defined structure and how to calculate the quantitative results presented with the KID.
Given the prescribed nature of a lot of the content, most firms will be able to meet the majority of the requirements using their existing knowledge of their business and their products.
However, the calculation of the “Market Risk Measure” and “Performance Scenarios” are likely to be challenging for smaller firms. For standard unit-linked funds (classed as Category 2), formulae are prescribed based on historic returns of either the fund or appropriate proxies. For with-profits funds (classed as Category 4), there is a requirement to use stochastic projections using a “boot-strapping” approach. To complicate this even further, firms need to take into account non-market observable items, such as smoothing and estate distribution. These are best tackled using an asset share approach to the projections.
The Market Risk Measure and Performance Scenario elements of the KIDs requires firms with with-profits products to describe the likely distribution of their product’s performance over its recommended holding period.
Boot-strapping enables firms to estimate this distribution given a limited sample of historic asset returns. For each simulation in the stochastic projection the boot-strapping technique requires you to randomly select a historic asset return for each projection period, noting that the same return can be selected multiple times.
After doing this for each projection period within each of the 10,000 simulations required by the regulations, and making adjustments to convert real-world observed returns into risk-free returns, firms will have enough data to feed into their asset-share models to calculate the product’s projected performance distribution.
"For many smaller firms this will be their first exposure to stochastic modelling, and possibly asset shares. "
For many smaller firms this will be their first exposure to stochastic modelling, and possibly asset shares. Those who were not subject to the Realistic Balance Sheet regime will have only had to calculate Cost of Guarantees (“CoGs”) for the first time under Solvency II regulations. Typically larger firms are calculating these CoGs using full stochastic models already. Our experience has been that smaller firms are calculating CoGs using approximate approaches such as the Black-Scholes formula or a simplistic scenario based approach on proportionate grounds. Unfortunately, neither of these approaches will meet KIDs requirements as a minimum of 10,000 simulations based on historic sampling is stipulated.
So how can Barnett Waddingham help you? We have produced a KIDs checklist tool and can advise on or review all aspects of producing compliant KIDS, including providing historic data for the use of fund proxies. In respect of the stochastic projections, we have a stochastic asset share model that enables us to calculate the Market Risk Measure and Performance Scenario outputs using the “boot-strapping” approach.
In most conditions, the Market Risk Measure and Performance scenarios will need to be re-calculated only annually. However, there is a need to monitor the market data and change the MRM if the most-common observed MRM value over the previous 4 months has changed. The frequency of monitoring is not specified but it appears to us that this will have to be at least monthly and so the MRM calculation will need to be calculated monthly. Whilst you may want to outsource the calculations to us, the most cost-effective approach might be for us to provide you with a version of the model for you to use and perform your own calculations and monitoring.
For further information on meeting the KIDs for PRIIPS requirements, please contact your usual Barnett Waddingham consultant.